TL;DR
Salesforce is acquiring m3ter, a London metering platform, to add native consumption billing to Agentforce Revenue Management.
Salesforce is acquiring m3ter, a London metering platform, to add native consumption billing to Agentforce Revenue Management.
Salesforce has signed a definitive agreement to acquire m3ter, a London-based metering and rating platform built for consumption-based billing. The deal will integrate m3ter’s infrastructure natively into Agentforce Revenue Management, giving Salesforce customers the ability to launch, track, and bill usage-based and outcome-based pricing models without leaving the platform. Financial terms were not disclosed.
The acquisition reflects a structural shift in how software companies charge for their products. Traditional per-seat subscriptions made sense when humans were the primary users, but AI agents that perform work autonomously create a billing problem: if one agent replaces ten employees, selling ten licences no longer works. Salesforce itself has been navigating this tension, moving Agentforce to a consumption model built on Flex Credits where each agent action costs roughly $0.10.
m3ter was founded in 2020 by Griffin Parry and John Griffin, who previously co-founded GameSparks, a cloud services company acquired by Amazon in 2017. The pair spent three years at AWS after the acquisition, where they saw first-hand how Amazon’s usage-based billing infrastructure worked at scale. They left to build m3ter as a standalone metering layer that could sit between a product and its billing system.
The platform ingests product usage data in near real time, applies configurable pricing rules, and outputs billable charges to whatever CRM, ERP, or invoicing system a company uses. m3ter raised $17.5 million in seed funding from Union Square Ventures, Insight Partners, and Kindred Capital in 2022, followed by a $14 million Series A led by Notion Capital in 2023. Its customers include Paddle, Onfido, and Sift.
“We founded m3ter to solve the hardest problems in usage-based pricing,” Parry said. “Joining Salesforce allows us to bring our high-scale mediation and rating capabilities to the world’s largest enterprise install base.” The transaction is expected to close in the second quarter of Salesforce’s fiscal year 2027, subject to customary closing conditions.
m3ter is the latest in a series of acquisitions Salesforce has made to assemble the infrastructure for its AI agent strategy. The company acquired Contentful earlier this month for a native content layer, completed an $8 billion deal for Informatica in late 2025 for data integration, and bought Momentum, Qualified, and Cimulate for conversation intelligence, AI sales engagement, and digital experience simulation respectively.
The pattern is clear: Salesforce is buying the components it needs to make Agentforce a complete platform rather than a feature bolted onto its existing CRM. m3ter fills the monetisation gap, the infrastructure required to actually charge customers for what AI agents do. Without native metering, enterprises running consumption-based models have to stitch together third-party billing tools or build custom integrations, a problem that becomes harder as pricing models grow more complex.
Whether this translates into revenue growth is the question investors are watching. Salesforce reported $11.13 billion in revenue for fiscal Q1 2027, up 13% year on year, and Agentforce reached $1.2 billion in annual recurring revenue. The stock fell roughly 1.7% on the day of the m3ter announcement, sitting closer to its 52-week low of $163.52 than its high of $276.80.
Investors want proof that consumption-based AI revenue can scale fast enough to offset the structural threat to seat-based licensing. A billing infrastructure acquisition is a bet on plumbing rather than a growth catalyst, and the market priced it accordingly.
For m3ter, the outcome is a fast exit for a company that raised just $31.5 million in total funding. For Salesforce, it is another piece in a stack that now spans data (Informatica), content (Contentful), agents (Agentforce), and billing (m3ter). The question is whether enterprises will consolidate on that stack or continue assembling their own from best-of-breed vendors, a choice that the shift to consumption pricing makes more consequential with every agent deployed.
Schools have been quietly chipping away at recess for nearly a decade, and a sweeping new update from the American Academy of Pediatrics says the consequences are real, measurable, and showing up well beyond elementary school. At the same time, the federal government has issued a formal advisory on children and screen time, calling on schools, parents, and tech companies to act. Both stories point in the same direction, but the path forward is far less obvious than the headlines suggest.
For the first time since 2013, the American Academy of Pediatrics has updated its recess guidelines, and the expansion is significant: the new recommendations extend to middle and high school students, not just younger children. EdSurge reporter Lauren Coffey has been reporting on what that guidance actually means for administrators under pressure to protect instructional time, whether the evidence on attendance and attention is strong enough to move policy, and why advocates say the answer may be simpler than schools are willing to admit.
A formal screen time advisory from HHS Secretary Robert F. Kennedy calls for bell-to-bell phone bans, warning labels on apps, and the elimination of recommendation algorithms for children. But researchers are being careful: the evidence linking screen time to negative outcomes is correlation, not proven cause and effect, and the line between harmful social media and beneficial education technology is one that schools and families are still figuring out how to draw. EdSurge reporter Nadia Tamez-Robledo breaks down what the advisory actually asks of schools, why the tech industry response will be the real test, and what the carve-outs for students with IEPs reveal about the limits of a one-size-fits-all approach.
Recess Took a Break in Some Schools. A Push Is On to Bring It Back by Lauren Coffey
Surgeon General Advisory Wants Kids to Live Beyond the Confines of Screens by Nadia Tamez-Robledo
Join us on This Week with EdSurge where we ask whether the research is pointing toward a simpler solution than most schools are willing to try. Listen to the episode.
This Week with EdSurge is produced by the EdSurge newsroom. Subscribe to the EdSurge newsletter for the latest in education news delivered straight to your inbox.

Smartphones deliver more power and polish than ever, yet most follow the same safe template. AYANEO decided to break the mold with its first smartphone. The Pocket Play takes the sliding concept from Sony’s long-gone Xperia Play and updates it for today’s games and apps. Slide the 6.8-inch display upward in landscape mode and the magic happens. A full set of physical controls appears underneath. You get a proper D-pad on the left, ABXY face buttons on the right, two round capacitive touchpads that stand in for analog sticks, plus shoulder bumpers and triggers, while dedicated shortcut buttons sit within easy reach.
The layout is similar to the 2011 Xperia Play, but with a more modern flair. Those spherical touchpads replace traditional analog sticks, yet they still provide a remarkably natural input experience, especially with the limited time you have to check them out. Keeping your thumbs off the screen allows you to have a continuous view even during the most intense gaming sessions. The entire control deck is pushed to the bottom, leaving the huge display completely unobstructed for whatever you’re doing, whether gaming or watching media.
Sale
The 6.8-inch display boasts a 2400 by 1080 resolution and a refresh rate of 165Hz. An OLED panel delivers stunning colors and silky smooth action, whether you’re playing high-frame-rate Android games or streaming from the cloud. When the device is closed, it acts like a phone for calls, messages, and app use; however, when you open it, it morphs into a dedicated gaming setup.
The MediaTek Dimensity 9300, along with its Immortalis-G720 graphics engine, provides excellent performance. AYANEO backs it up with lots of high-speed LPDDR5 RAM and UFS 4.0 storage. You also get a microSD card slot, which is a nice touch given how useful they are for adding new games or emulator roms to your library. The active cooling mechanism keeps everything operating smoothly, even during extended gaming sessions, which is unfortunately uncommon on most smartphones.
A 5,000 mAh battery keeps the lights on, and a quick recharge means you won’t have to wait long before your next gaming session. You also get stereo speakers and a strong vibration motor, which work together to create a really immersive gaming experience with excellent audio and haptic feedback. A USB-C port with speeds of up to 3.1 Gen 2 and a DisplayPort 1.4 output enable you to connect the device to a TV or monitor and enjoy games on the big screen. You also have Wi-Fi and Bluetooth to suit your wireless requirements.
The device’s cameras are mounted on the back and include a 50-megapixel primary sensor and a 16-megapixel ultrawide lens. There is a 5-megapixel selfie camera on the front. To be fair, these cameras exist merely to handle the basics and are unlikely to impress anyone; early reports show they do the job, but don’t expect high-end photography from them. It’s clear that AYANEO emphasized gaming gear over taking beautiful photos.
You get Android 15 out of the box, and a fingerprint scanner is buried inside the power button for quick unlock. The software offers a full smartphone experience in addition to the typical gaming capabilities. There are also some useful shortcut buttons and a slide mechanism that allow you to quickly access game controls without having to deal with on-screen overlays.Pricing is unknown until the Kickstarter campaign opens. Rumors suggest that early backers will pay around $500, with future tiers perhaps providing extra RAM, storage, or higher-end finishes. AYANEO also intends to offer additional grips and cases to improve comfort for both gaming and non-gaming applications.
You can spend a whole lotta money on a good office chair these days – but however feature-rich and overly engineered they are, in my experience, you can’t beat a comfortable, solid office chair that won’t break the bank.
Of all the office chairs I’ve tested and reviewed, I still think the $210 Boulies EP200 on Amazon is the best value chair you can get right now. For me, the EP200, which is also on sale in the UK for £190 at Amazon delivers all the core features I expect, and more. Amazon has matched the sale price of this chair, which is discounted at both Boulies.com and Boulies.co.uk. It amounts to around a 30% discount, whichever site you choose.
I’ve been sitting on this chair day in, day out for 18 months straight now and it’s as good today as it was when I first assembled it. The mesh is still firm, supportive, and comfortable. There’s no creaks, no squeaks. If I were buying a new office chair today, I’d just get another EP200. It’s worth every penny. It scored 4.5 stars in my review, with a Highly Recommended award.
I absolutely love the EP200. I love the breathable mesh design, the lockable recline, the adjustable seat depth, even the 3D armrests with their satisfying click on every twist. I use this at work every day, but also after-hours when I’m gaming. I never feel discomfort or fatigue that some cheap chairs induce (and some expensive ones, for that matter).
Now, Boulies’ value-driven chair isn’t as rich with features as the Herman Miller and Steelcase office chairs my team and I have tested. But it’s well-built for the money, highly adjustable, and doesn’t require taking out a bank loan to afford it.
Having spent many hours in an office sitting on the iconic Herman Miller Aeron, I can tell you that while that pricey mesh chair was very comfortable, for most people working in an office or home office, the EP200 delivers everything you’d want from a seat at a fraction of the price.
In summarizing my time with the EP200 in my review update, I said “Overall, a year on, I still find the Boulies EP200 to be the archetypal office chair for most people. It’s relatively cheap, not overly engineered or designed (in a good way), and suitable for long hours in the office and home office.”
For more top-performing options, check out my in-depth guide to the best office chairs we’ve tested (and yep, the EP200 is on that list).
Industry 4.0 represents an opportunity for innovative, ambitious and future-focused professionals to transform the world as we know it.
The fourth industrial revolution, or Industry 4.0 as it is often better known, is the integration of smart, digital technologies into larger industrial and manufacturing processes, with the aim of creating intelligent, improved systems.
Skills in this area are absolutely vital for any professionals looking to work in a future-focused role or high-tech capacity.
SiliconRepublic.com, as part of its month-long Industry 4.0 coverage, has compiled a list of some of the most important skills to have as you face a changing world.
It is not at all uncommon to assume that the most critical qualities to possess in technical and complex industries are all hard skills. It is, however, a commonly held myth, as soft skills are critical to long-term success in any professional space, including for careers under the Industry 4.0 umbrella. These are the abilities that enable you to communicate effectively with co-workers, negotiate positive change and create a better workplace environment.
Skills to prioritise are adaptability and problem-solving, as Industry 4.0 roles are often complex and ever-evolving; collaboration, as your job may demand a degree of crossover with other teams, departments or companies; critical thinking, as Industry 4.0 careers are often rooted in a need to address modern-day problems with unique solutions; and leadership, as everyone should know how to command a room and lead others when necessary.
Soft skills bring a crucial human element to careers that are often considered complex and clinical.
3D printing – or as it is often known, additive manufacturing – has enabled experts in Industry 4.0 careers to move beyond traditional methods as they create models and prototypes with the power to improve quality of life, reduce costs and maximise resources.
Clinicians may use 3D printing to develop prosthetics and implants to match a patient’s specific anatomy. Manufacturers can use the technology to create specific, necessary medical equipment and aerospace engineers often use additive manufacturing to develop small, complex parts that demand high accuracy and specific criteria.
There are a range of organisations and sectors in STEM and outside of it that are now using 3D advancements and it is undoubtedly a skill that will be carried into the next industrial wave.
Digital twin technology is described as the virtual representation of a physical system or process that receives data from the real world, in real-time. Its purpose is to mirror the behaviour, performance and state of the primary, physical model, so experts can explore, experiment and analyse without impacting the real counterpart.
Digital twin tech allows organisations to simulate real-world scenarios, fortify security and improve operations, while also minimising risk or accidental harm.
Among the skills that are useful to those who want to know more about wielding digital twin technologies are abilities in IoT, AI, data analytics, simulation software and cloud computing – and it is of particular importance to those hoping to work in manufacturing, aerospace, automotive, energy and utilities, healthcare, smart cities, and infrastructure.
Research suggests that IoT, as a critical element of Industry 4.0, is a skill that is under near constant demand by organisations and employers.
Anyone hoping to be an IoT expert should ensure that they have a robust education in software such as AutoCad, which allows engineers to design machines; cybersecurity for managing complex and high-risk projects; data and analytics skills that ensure clean, concise and insightful work; and AI and ML, among others.
Careers in Industry 4.0 are moving so rapidly it can be hard to keep up with the changes, but what is important to remember is that the skills of today create the future. All you have to do to keep pace is to commit to upskilling and tackle each challenge as it comes.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
The JDY botnet, a malware network previously associated with Chinese threat actors like Volt Typhoon, has significantly expanded its targeting scope and reconnaissance efforts.
According to researchers at Black Lotus Labs by Lumen, who have been monitoring its activity, JDY maintains a strong focus on the United States, where many of its compromised devices are located and where it heavily targets military and associated networks.
The security firm notes that JDY has grown from roughly 650 active bots in January 2024 to over 1,500 compromised SOHO and IoT devices today.
While the numbers seem low, it’s important to note that JDY isn’t an exploitation framework or a DDoS botnet that requires large swarms to accumulate firepower, but is instead a distributed scanning and fingerprinting network that helps its operators locate targets vulnerable to newly disclosed flaws.
“Analysis of this activity shows a clear focus on identifying vulnerable infrastructure shortly after public vulnerability disclosures, suggesting that reconnaissance output is rapidly operationalized by China-nexus advanced persistent threat (APT) actors,” reads the Black Lotus Labs report.
“This targeted focus has been observed across a range of sectors, with the U.S. military and associated entities as the most prominent.”

CISA has previously warned about the risk Volt Typhoon operatives pose to unprotected SOHO routers, urging network device vendors to eliminate vulnerabilities in SOHO router web management interfaces (WMIs) during the design and development phases.
The JDY botnet is designed to conduct service discovery, service banner grabbing, TLS certificate collection, protocol fingerprinting, and flaw-focused reconnaissance.
Among the compromised devices are those from Cisco, Araknis, Mimosa Networks, Ubiquiti, DrayTek, Hikvision, and Linksys, for MIPS, MIPS64, MIPSEL, and MIPSEL64 architectures.
The threat actors are quick to target newly disclosed vulnerabilities, with Lumen researchers observing JDY scans targeting CVE-2026-35616 shortly after Fortinet publicly disclosed the FortiClient EMS flaw.

The operators control the botnet through hidden Tor services, which also serve as command-and-control (C2) infrastructure. The open-source reverse-shell and host-management framework Platypus is also used in some cases.

The malware registers with a central “Dispatch Service” and receives scanning assignments, which it executes, compresses the results, and sends them back to the C2.
The scanning module supports the following:
The botnet client repeats the same cycle until the operator specifically orders it to stop.
The TCP scanning function is one of the most technically interesting, say the researchers, explaining that, when JDY has sufficient privileges, it performs much faster and stealthier raw SYN scanning.
“If the malware can open a raw socket, which generally requires root or administrative privileges, it initiates high-speed SYN scanning using custom-crafted TCP packets,” explains the report.
“These custom packets use a fixed source port of 19000, increment the destination ports one at a time, and batch-process thousands of scan targets.”

As JDY botnet activity increases, organizations should ensure routers, firewalls, and IoT devices are running the latest security updates and patches to prevent them from being recruited into reconnaissance networks.
Defenders should also reduce their external attack surface by disabling unnecessary internet-exposed administrative interfaces, restricting remote management access, replacing default credentials, and monitoring for unusual outbound scanning activity originating from edge devices.
Security teams log 54% of successful attacks and alert on just 14%. The rest move through your environment unseen.
The Picus whitepaper shows how breach and attack simulation tests your SIEM and EDR rules so threats stop slipping by detection.
iPhones are often easiest to recommend if you’re looking for a device that just works, with software that’s reliable. That said, iOS 26’s many issues pushed Apple to finally focus on performance and stability with its next big release. At WWDC 2026, the company unveiled iOS 27, which is set to roll out to the masses in September. If you’re keen to try out the new update, however, you can always install the beta build on your iPhone. Public betas are often more stable, while developer beta builds tend to get all the new features earlier.
Apple no longer requires you to pay a fee to test its beta builds, as it did a couple of years ago. All you need to do is head to the Apple Beta website and sign in using your Apple ID and password. On the terms and conditions page, click on “Agree.” Restart your iPhone and navigate to Settings > General > Software Update, and you should now be able to see a new “Beta Updates” section. Tap on it and select “iOS 27 Developer Beta.” Give it a quick second, and your iPhone should now let you download and install the newest beta version of iOS.
Depending on your internet connection, the process may take anywhere from a few minutes to an hour or more. While you won’t lose any data, it’s highly recommended you make a backup of your iPhone prior to installing any beta builds.
Apple seldom makes major changes with its products, so when iOS 26 got a facelift with the Liquid Glass design system last year, everyone was excited. It did come at the cost of iOS 26 slowing down iPhones, making iOS 27’s emphasis on performance and stability improvements its biggest saving grace. We’ve been testing the first developer build of iOS 27 and can confirm that it already feels snappier than iOS 26 ever did — but this isn’t a good enough reason for everyone to rush out and install the beta.
Though you get to try out the cool new things, including Siri AI, a developer beta is meant primarily for testing purposes. We do not recommend installing a preview build on your iPhone if it’s your primary device, since there’s a chance that things may break or not function as they’re supposed to. It’s also worth mentioning that the hallmark feature — Siri AI — seems to be rolling out in phases, with many early testers still stuck on a waitlist.
Battery life is also almost always poor with early beta builds, so that’s another thing to keep in mind. It often takes a few developer beta updates to get the noticeable bugs ironed out. Apple usually releases the public beta version by the end of July, which should be considerably more stable if you’re still keen to try iOS 27 out before September.
If you look around your environment, you can probably pick off quite a few things that you’ve made, at least if you’ve been at this a while. You probably aren’t reading this from the bottom of a body of water though, which means you lack the incredible confidence of submarine builder [Hank Pronk]. Not only is he building himself a capable-looking diesel-electric submarine over on YouTube, he’s even DIYing CO2 scrubbers for it! Yeah, that’s a man who believes in himself.
Luckily [Hank] is not anywhere near the Caribbean, so needn’t worry about being misidentified as a narco-sub, but he still has to be concerned about his oxygen supply when tooling around beneath the local lakes. Perhaps more important than the oxygen supply in a sub is the build up of CO2. It doesn’t matter how many oxygen tanks you bring down with you if you can’t scrub CO2 out of the air to make room for it. Just like the Apollo missions, he’s using a chemical adsorbent to take carbon dioxide out of the air — and just like Apollo 13, he’s switching from square to round.
Or, rather, from a rather rectangular commercial model to a DIY little round unit. That’s because he doesn’t need the big scrubber in this sub: being diesel-powered, he expects to spend a lot of time at snorkel depth, where both the pilot and the engines can get clean air through the tube. Dives are expected to be short, and in that use case, too big of a CO2 scrubber is really a waste. If for some reason he gets stuck on the bottom, well, the lake isn’t that deep. He can swim to surface, and has a detailed bailout plan. If he wants to stay under overnight to avoid bailing at night, he’s carrying enough extra adsorbent for that.
There’s a reason almost every submarine we’ve featured on this site over the years is an ROV. It’s not that a homemade submarine is automatically a death trap, but you sure do have to be confident in your design.
Most of us spend more time wearing headphones or earbuds than we realize, and I know I do. Between shuffling playlists during my daily commute, jumping on work calls, listening to podcasts while walking around the neighborhood, and winding down with a late-night binge session, audio has quietly become an essential part of everyday life.
Our listening needs, however, change throughout the day. The over-ear headphones I want for a long workday are not necessarily the same audio wearable I reach for when stepping outside. Yet many audio brands seem determined to convince us that a single expensive product is the ultimate solution.
In an era where new wireless gear launches every other week at eye-watering price tags, finding tech that genuinely balances comfort, performance, and value can feel like a challenge. That is exactly what drew me to JLab. Rather than relying on exclusivity or celebrity endorsements, the brand has built its reputation on creating practical tech designed for real people and real-world routines.
This approach is incredibly refreshing. By delivering high-tier features at an accessible price point, JLab leaves a lasting impression and emerges as a compelling choice for buyers. You don’t have to nuke your wallet to enjoy great sound in your everyday life. That idea has been distilled to the very core of two products in particular — the JBuds Lux ANC Wireless Headphones and the Epic Pods ANC True Wireless Earbuds.

There are times when I want to tune everything else out. Whether I am trying to focus on work in a buzzing cafe, settling into a long flight, or simply listening to a new album without distractions, there are times when noise isolation is as important as the raw depth of tunes blasting into your ear canals. This is where over-ear headphones often come to the rescue, offering a level of immersion that earbuds cannot quite match. The JBuds Lux ANC Wireless Headphones embody that scenario, and they have been accordingly designed to make those moments feel even more effortless.
Featuring Smart Active Noise Cancelling, they help reduce background distractions so you can stay focused on your music, podcasts, movies, or work calls rather than what is happening around you. The convenience extends beyond the bliss of noise cancellation. Bluetooth Multipoint connectivity allows you to stay connected to both your laptop and phone at the same time. You can seamlessly switch from a Zoom call on your computer to a playlist on your phone without ever needing to re-pair your devices.
That same focus on practicality carries over to the per-charge battery mileage, as well. With more than 70 hours of playtime, these headphones are built to keep pace with packed schedules, extended travel, and marathon listening sessions without the need for frequent recharging.
Just as important as battery life is how the headphones feel after hours of use. The over-ear design and Cloud Foam cushions are made for extended wear, making them an easy companion for long workdays, study sessions, and travel. Heading into the JLab App, users can also fine-tune their sound settings to match their listening preferences, helping ensure every playlist, podcast, and movie sounds just the way they like it.

Of course, life does not always happen at a desk. Some days are a constant cycle of commuting, unexpected errands, out-of-schedule workouts, and quick transitions between one activity and the next. Those are the moments when portability becomes just as important as performance.
The Epic Pods ANC True Wireless Earbuds are designed for the aforesaid kind of on-the-go lifestyle. Built for people who are constantly moving, they pack an impressive amount of technology into a compact form factor. Instead of relying on a single driver to do all the work, their hybrid dual-driver system handles audio output with a level of depth and clarity that is hard to find in this segment. A 10mm dynamic driver handles deep bass, while a specialized Knowles Balanced Armature focuses on crisp mids and clear highs. The result is a balanced listening experience where vocals and instruments remain clear without getting overshadowed by the low end.
If great sound matters to you, support for LDAC on Android and AAC on iOS ensures that your music comes through with greater detail, clarity, and depth, especially when streaming high-quality audio. The Epic Pods ANC also feature Adaptive Active Noise Cancelling that goes beyond simply switching on and off. From crowded subway commutes to quieter streets, the earbuds adjust noise cancellation on the fly, helping maintain the right balance between immersion and awareness without you having to tap a button.
But impressive audio performance means little if your earbuds cannot keep up with your active lifestyle. With more than 56 hours of total playtime and a quick charge feature that delivers up to 5 hours of listening from just a 10-minute charge, the earbuds are built to keep up with even the busiest schedules.
The Epic Pods ANC are designed with everyday use in mind. The secure fit design helps keep them comfortably in place during intense workouts, while IP55-tier ingress protection provides added protection against sweat and dust exposure. Bluetooth Multipoint further adds even more convenience by allowing seamless switching between multiple connected devices.
The biggest shift happening in personal audio right now isn’t just about pushing the boundaries for better sound quality, but rather about changing expectations and offering a holistic package that is practically rewarding in more ways than one. Audio wearable enthusiasts no longer want to feel forced into choosing between premium features and a reasonable price tag. Instead, they want gear that fits naturally into our daily routines, solves those annoying everyday frustrations, and delivers genuine value.
This is what makes JLab a compelling choice. Whether you prefer the immersive comfort and endurance of the JBuds Lux ANC Wireless Headphones or the flexibility of the Epic Pods ANC True Wireless Earbuds, both models reflect the very same philosophy. Personal tech should always adapt to your life, not the other way around. And while at it, great audio shouldn’t be reserved for just a few buyers. It should be accessible, practical, and ready for whatever your day has in store.

The interests of Microsoft and graduates rebelling against AI are actually aligned.
That was one takeaway for Brad Smith, Microsoft president and vice chair, from a recent return to his alma mater, Princeton University, for its reunion weekend. Seniors wore class jackets labeled “100 percent cotton” and “100 percent human,” referencing allegations that an earlier design was created with AI — part of a broader backlash across campuses this spring.
In a blog post this morning, which he started drafting during that visit, Smith writes that graduates booing AI at commencements across the country are “telling us what we need to hear. He points out that Microsoft’s own future depends on people staying employed.
“Workers have been Microsoft’s lifeblood from the start,” he writes in the post. “If the world’s people don’t have jobs, then neither do we. And if we’re not doing our part to help people use technology to pursue better jobs, then we’re not doing the job we were born to do.”
Speaking with GeekWire this week, Smith acknowledged the tension between that message and job cuts across the tech sector, including at Microsoft. He addressed the issue in the post, as well, citing the industry’s desire to offset capital spending on AI, along with factors including geopolitical uncertainty, trade tensions, and a correction from earlier over-hiring.
“Our industry is going through one of the most extraordinary transformations in its history,” Smith said in the interview, while adding that the “expenses of capital expansion make it more difficult to afford the employment bubbles we’ve had, especially since 2020.”
Smith cited the automation of entry-level tasks among the challenges facing graduates, as well.
But he also took a larger view. Computer science jobs are changing, he said, not vanishing. Coding is becoming a smaller part of the work, while the roles around it — including designing software, managing product development, and reviewing code — are expanding.
In the post, Smith places AI in a longer line of technologies that reshaped work without ending it, from the camera to the spreadsheet to email. He calls AI the next “general purpose technology,” akin to electricity, and argues its spread will take decades, not years, because the limit is how fast people and institutions change, not how fast the models improve.
Some jobs go away, he writes, while new ones appear, and many are remade.
Smith’s advice to workers is to treat a job as a bundle of tasks rather than a title, sorting them into what AI can do, what a person can do with AI, and what only a human can do. For this, he takes inspiration from a new book by LinkedIn’s Ryan Roslansky and Aneesh Raman, “Open to Work,” and its list of durable human attributes: curiosity, creativity, compassion, communications, and courage.
The post also offers a clear message for companies, aligning with Microsoft’s own business interests. Smith says organizations need to build their own AI systems on top of frontier models, using their own data and what Microsoft CEO Satya Nadella calls a “hill climbing machine” of evaluations and steady improvement, rather than simply renting intelligence from someone else.
Smith cites intellectual property and data sovereignty as a central concern, arguing that firms must adopt AI without handing their hard-won expertise to a rival’s model.
In the interview, Smith said the blog reflects months of discussion among Microsoft’s senior leaders, including Nadella and Chief People Officer Amy Coleman, and that it’s intended to speak to the company’s own employees as much as to the outside world.
Asked what he would have told new college graduates had he been the speaker at a commencement ceremony this spring, Smith said he would have focused on the resilience of humanity more than advances in technology — urging them to speak up for the values they care about, help contribute to a better world, and go forward with hope and optimism.
“That doesn’t mean these challenges may not be significant,” he said, “but I personally believe that the human spirit is far greater than any artificial intelligence the world is likely to create.”
This article is brought to you by AGILINK.
Throughout the exhibition hall at the 2026 IEEE International Conference on Robotics (ICRA), in Vienna, one demonstration seemed to attract a disproportionate amount of attention.
Two robotic hands were making a balloon dog. Slowly and deliberately, the robot twisted a long balloon into loops, bends, and joints without popping it. Visitors stopped, watched, and often returned with colleagues to watch again.
AGILINK’s balloon dog demonstration draws a crowd at ICRA 2026.AGILINK
At first glance, the demonstration appeared almost playful. Among roboticists, however, balloon twisting is widely recognized as an unusually difficult manipulation task.
A balloon is lightweight, highly deformable, slippery, and extremely sensitive to force. Every twist changes its geometry and internal pressure, turning a seemingly simple activity into a continuously changing physical interaction problem.
Humans navigate those changes almost intuitively. While making a balloon animal, people rarely think consciously about force regulation, slip prevention, or contact stability. They simply adjust.
For robots, those adjustments remain remarkably difficult. The challenge is not merely moving fingers to the right positions. The harder part is maintaining stable interaction while the object itself is changing.
Highlights from AGILINK’s ICRA 2026 demonstrations, including visuotactile sensing, in-hand manipulation, balloon-animal shaping, and other contact-rich tasks enabled by the company’s latest OmniHand platform.AGILINK
That distinction helps explain why the balloon dog drew so much attention in Vienna. What appeared to be a dexterity demonstration was, in many ways, a demonstration about contact itself.
As robotic manipulation continues to advance, a growing number of researchers are arriving at a similar conclusion: many of the hardest problems in robotics begin only after contact occurs.
Balloon twisting combines two challenges that robotics has traditionally struggled to solve simultaneously: long-horizon task execution and contact-rich manipulation.
The first concerns motion.
A balloon dog is not created through a single grasp or twist. It emerges through a carefully ordered sequence of manipulations, each setting the conditions for what follows. A small rotational error introduced early may appear insignificant at first, yet several steps later it can prevent the final structure from forming altogether.
In that sense, balloon twisting is a long-horizon task. Success depends not only on performing individual actions correctly, but also on preserving the future feasibility of the entire manipulation process.
To address this challenge, AGILINK began by collecting demonstrations from professional balloon artists. Human actions were mapped onto robotic hands to establish an initial manipulation policy. But successful demonstrations alone were insufficient.
In practice, some of the most valuable learning occurred when execution began to drift toward failure. Whenever instability emerged, human operators intervened and corrected the manipulation in real time. Those interventions were recorded and incorporated into reinforcement-learning cycles, allowing the system to learn not only how successful demonstrations unfold, but also how experienced operators recover when things start to go wrong.
Through this process, the robot gradually acquired the capabilities required for long-horizon task execution—a collection of abilities that AGILINK groups under the term motion intelligence: the ability to generate actions, coordinate bimanual behaviors, and execute extended manipulation sequences under real-world uncertainty.
OmniHand 3 Ultra-M on display at ICRA 2026.AGILINK
Yet motion alone does not explain why balloon twisting remains difficult. The second challenge is contact.
The robot must continuously regulate force, adjust contact locations, and respond to subtle changes in the object’s state. These decisions are difficult to encode through explicit rules. Even skilled human operators often rely on tactile intuition developed through experience rather than consciously articulated strategies.
Analysis of those interventions revealed that many failures did not originate from incorrect action sequences, but from the breakdown of contact itself.
To better capture those interaction dynamics, AGILINK collected contact-centric intervention data and incorporated those interactions into reinforcement-learning training. Rather than learning only which motions to perform, the system also learned how humans maintain stability when contact conditions begin to deteriorate.
AGILINK describes this capability as contact intelligence: the ability to establish, maintain, and adapt physical interaction as force distribution, friction, deformation, and contact geometry continuously evolve.
The distinction between the two capabilities is subtle but important. Motion intelligence determines what the robot intends to do. Contact intelligence determines whether it can continue doing it. For balloon twisting, both are necessary. One provides the sequence of actions. The other keeps those actions physically viable.
YouTuber KhanFlicks follows OmniHand’s motions while learning to fold a balloon dog at the AGILINK booth.AGILINK
Between a balloon slipping away and a balloon bursting lies a narrow region of stability. Successful manipulation depends on finding that region—and remaining within it throughout the task.
The balloon dog demonstration showcased a manipulation capability. It also revealed a broader question. How much contact intelligence can be achieved through learning alone? A robot can only regulate what it can perceive. It can only respond as quickly as its hardware allows.
As manipulation tasks become increasingly complex, researchers are finding that progress depends not only on better policies, but also on richer sensing and faster physical response.
That realization formed the backdrop for AGILINK’s second major announcement at ICRA 2026. Alongside the balloon dog demonstration, the company introduced the OmniHand 3 Ultra-M.
OmniHand 3 Ultra-M closely matches the size of an adult human hand.AGILINK
The two exhibits represented different stages of the same technological trajectory. If the balloon dog demonstrated what contact intelligence can already accomplish today, Ultra-M was designed to explore what contact intelligence may require next.
Roughly the size of an adult human hand, the OmniHand 3 Ultra-M integrates 20 active degrees of freedom within a human-scale form factor.
Its most distinctive feature is a fully direct-drive architecture. By adopting direct-drive actuation throughout the system, the hand is designed to enable faster and more transparent force regulation and higher force-control bandwidth, enabling faster response as contact conditions change. For contact-rich manipulation, responsiveness can be as important as sensing itself.
By adopting direct-drive actuation throughout the system, the OmniHand 3 Ultra-M is designed to enable faster and more transparent force regulation and higher force-control bandwidth, enabling faster response as contact conditions change.
The platform also incorporates tactile sensing across nearly the entire hand. Each fingertip contains a miniature vision-based tactile sensor, while more than 300 three-dimensional tactile sensing points are distributed throughout the palm. Together, they provide information not only about where contact occurs, but how contact is evolving.
The system is designed to estimate pressure distribution, shear forces, local deformation, slip tendencies, and other interaction dynamics that often remain invisible to conventional position-based control systems.
According to AGILINK’s tests, individual sensors achieve force resolution of approximately 0.005 N—roughly equivalent to detecting the weight of a sheet of paper resting on a fingertip. Spatial resolution reaches approximately 0.04 mm, while sensing density approaches 50,000 sensing points per square centimeter.
OmniHand 3 Ultra-M recognizes feather texture through vision-based tactile sensing.AGILINK
For dexterous robots, contact has traditionally been a largely hidden process. Ultra-M is designed to make that process more observable.
Rather than simply detecting that contact has occurred, the system attempts to resolve where interaction is happening, how forces are distributed, whether instability is beginning to emerge, and how manipulation strategies should adapt in response.
The balloon dog offered a glimpse of what contact intelligence can already accomplish. Ultra-M explores a different question: what capabilities may be required to push contact intelligence further?
The significance of contact intelligence extends far beyond balloon animals. Many tasks that continue to resist automation involve unstable or deformable interaction: cable insertion, garment handling, flexible packaging, delicate assembly, connector mating, tool use, and household manipulation.
These tasks are difficult not because robots cannot reach the correct location, but because maintaining stable interaction after contact begins remains extraordinarily hard.
For decades, robotics achieved many of its successes by reducing uncertainty. Factories were engineered to make robotic motion predictable, repeatable, and highly structured. The physical world behaves differently.
A growing share of robotics research is shifting toward interaction itself—understanding how robots can establish, maintain, and adapt physical contact within environments that remain fundamentally unpredictable.
Objects shift. Materials deform. Friction changes. Contact evolves. Real environments rarely follow scripts. Seen through that lens, the balloon dog was never really about the balloon dog. What attracted attention at ICRA was not simply a visually impressive demonstration, but what it revealed: intelligence in the physical world is ultimately measured through interaction.
As motion generation continues to mature, a growing share of robotics research is shifting toward interaction itself—understanding how robots can establish, maintain, and adapt physical contact within environments that remain fundamentally unpredictable.
For robots moving beyond structured environments and into less predictable real-world settings, managing contact may become as important as motion itself.
Weekend Open Thread: Evereve – Corporette.com
Jensen Huang Approves Samsung, SK Hynix, and Micron for NVIDIA (NVDA) HBM4 Memory Supply
Anatomy of the June crypto crash: Fed, Iran, Saylor
The Best Mystery Series of All Time Is Surging on Streaming 30 Years After It Ended
Alexander Zverev wins the French Open to finally earn a 1st Grand Slam title
Suspicious Polyfill login prompts pop up on Toshiba, Muji websites
Senator Cynthia Lummis Calls CLARITY Act the Most Consequential Financial Legislation of This Generation
Microsoft unveils seven homegrown AI models in new bid for ‘long term self-sufficiency’
Microsoft launches MXC, an OS-level sandbox for AI agents, with OpenAI and Nvidia already on board
(VIDEO) Justin Bieber Delivers Surprise Happy Birthday Serenade to Diners at Los Angeles Mexican Restaurant
The Pain Points Taking a Fragile Tech Rally Down a Notch
LBank Surpasses 25 Million Users Worldwide as AFA Partnership Continues to Drive Global Growth
Trump’s AI Ownership Plan Could Benefit Anthropic at OpenAI’s Expense
Eli Lilly (LLY) Stock Surges 4% Following Breakthrough Sleep Apnea Trial Results
Meta steals a tactic from Tesla and builds data centers in tents
Von der Leyen’s AI envoy pick draws conflict-of-interest fire
Bangladesh beat Australia after 20 years in ODIs, register only their second win over six-time world champions | Cricket News
High Stakes for Wembanyama as New York Pushes for 3-0 Lead
Hackers now exploit SolarWinds Serv-U flaw to crash servers
Notion restores access to Anthropic after service disruption
You must be logged in to post a comment Login